JP3520159B2 - Thread unevenness detection method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting yarn unevenness that appears as periodic fluctuations of spun yarn.

[0002]

2. Description of the Related Art As a device for detecting yarn unevenness of spun yarn,
The Applicant offers the trade name Evenness Tester. In this conventional technique, a yarn sample is made to run between the electrodes, and a signal proportional to the weight change of the yarn is taken out.
For example, the width of the electrode in the running direction of the thread sample is 8 mm
At this time, if the sample is run at 400 m / min, the measurement length will be 2 km in the test of 5 minutes. If this signal is recorded according to time, a weight diagram is obtained, and if a distribution of amplitude is taken, a weight histogram is obtained. Also, if CV% and U% are calculated, it becomes the variation rate of the average weight of 8 mm length.

By the way, as shown in FIG. 1 (A), the periodic fluctuation of the thread unevenness is not limited to a complete sine wave, but generally, a waveform greatly distorted from a sine wave, a triangular wave shape, a rectangular wave shape, etc. In the waveform of, the fluctuation having periodicity is included. In addition to the fundamental wave (wavelength λ), these have wavelengths λ / 2 and λ /
3 and λ / 4, which appear as a harmonic of a whole number. For example, analyzing the waveform shown by the solid line in FIG. 1A, in addition to the fundamental wave A (wavelength λ), the harmonic wave B (wavelength λ /
2) and the harmonic C (wavelength λ / 3) are included.

Therefore, if a signal taken out from one yarn sample and proportional to the weight fluctuation is frequency-resolved (or wavelength-resolved) by Fourier transform, a spectrogram as shown in FIG. 1B (hereinafter referred to as an individual spectrogram) is obtained. Can be formed. In the above-mentioned product name, the Evenness Tester, the horizontal axis is the wavelength, and the doubled wavelength is decomposed into five, with the short side on the left and the long side on the right.
It is displayed on a logarithmic scale. That is, the distance between one wavelength λ ₁ in the yarn unevenness component and the wavelength λ ₁ × 2 which is twice the wavelength is 5
It is disassembled into two and each of these disassembled parts is a "channel".
I am trying. The height of each channel is the variation rate of the yarn unevenness component contained in the channel, and by analyzing this, the mechanical defect portion in the spinning process can be detected. For example, in the case of FIG. 1B, this indicates that the rotor in the open-end spun yarn has a defect, and unevenness (wavelength 20c
m (second) (wavelength 10 cm), third (wavelength 6.7c)
m), the channel showing the fourth harmonic (wavelength 5 cm) is protruding. When the square of the value of each channel is added for all channels and the square root is calculated, the above CV
% And U%.

[0005]

As described above, since the individual spectrogram shows the fluctuation rate of the yarn unevenness component for each channel, it is possible to analyze the fluctuation ratio of the yarn unevenness component to make a defect in machine adjustment and parts in the spinning process. Or it can detect if there is a defect.

However, such an analysis is currently carried out by relying on the experience of an experienced engineer.
In recent years, when the number of skilled technicians is decreasing, it is desired to make the analysis of the channel as simple and easy as possible.

With respect to this point, the present inventor theoretically calculates the yarn unevenness variation rate of an ideal spun yarn and expresses it as an ideal spectrum, and when compared with the individual spectrogram, among the individual spectrograms, We have found that channels that exceed the ideal spectrum can be easily found at a glance. This ideal spectrum can be theoretically determined if the average thickness of the fibers that make up the yarn and the length distribution of the fibers (the approximate average fiber length is also acceptable) are known. The thickness can be measured by a device (trade name: Micronaire) provided by the applicant, and the distribution of fiber lengths or the average fiber length can be measured by a device (trade name: Krashi fiber) provided by the applicant.

For example, FIG. 1C shows an individual spectrogram to which an ideal spectrum curve is added. Therefore, in this case, by comparing the ideal spectrum with the individual spectrogram, it is possible to easily confirm the periodic defect of the sample from the channel of the individual spectrogram that significantly exceeds the ideal spectrum.

By the way, as shown in the figure, in the individual spectrogram obtained from the actual sample, the channel height is always higher than the ideal spectrum in all channels (referred to as I index). This is because there is unevenness (additional unevenness) that is added due to the influence of the movement of the machine in the spinning process. The higher the adjustment or the defective parts in the machine, the higher the height. In the spinning process, defects and additional unevenness in the initial stage are stretched and become defects and unevenness with a long wavelength in the final stage of the spinning process.Therefore, by checking the wavelength and height of additional unevenness in the individual spectrogram, which process Can determine if there is a problem. That is, in the individual spectrogram, it can be determined that there is no problem in the channel within the allowable I index range, but when the specific channel is significantly higher than the preceding and following channels, for example, 1.5.
If it is as high as twice or more, it means that the specific period unevenness component exceeds the range of the allowable I index, and it is judged that there is a mechanical defect of a period corresponding to that.

However, in the case of the method of comparing the ideal spectrum and the individual spectrogram, whether or not the height displayed in each channel as described above should be regarded as a defect in the spinning process of the yarn,
Furthermore, judgment based on experience is required. However, as mentioned above, the individual spectrogram obtained from the sample is displayed at a height exceeding the ideal spectrum over all channels, so is it within the allowable I-index range that should not be regarded as a defect? It is necessary to make a judgment. Further, the yarns supplied to the market vary depending on grades and the like. For example, in the case of high quality yarns, the I index must be strictly judged, and in the case of low quality yarns, the I index must be judged generously. It is not possible, so it involves difficult decisions.

[0011]

According to the present invention, of the unevenness components contained in the yarn to be measured, it is judged whether or not the additional unevenness that does not need to be regarded as a defect in the yarn is based on the defect. The present invention provides a method for detecting unevenness in a yarn, in which only important defective unevenness to be seen as a defect can be easily found at a glance in the yarn.

Therefore, the first aspect of the present invention is that a plurality of samples sampled from the yarn to be measured are run, a signal proportional to the weight fluctuation of each sample is extracted, and then the signal is extracted from the signals. In the yarn unevenness detection method, the yarn unevenness components having different long and short wavelengths are classified for each channel by frequency analysis or wavelength analysis, and the variation rate of each yarn unevenness component is displayed by the height of the channel to form an individual spectrogram. To form an overall spectrogram by averaging the k individual spectrograms obtained corresponding to, and a reference spectrogram is formed by averaging the heights of adjacent channels of the overall spectrogram. Spectrogram and Overall Spectroscopy By comparing the ram, to display the height ultra excessive portion of the channel of which exceed the reference spectrogram of overall spectrogram, thereby in that it detects a yarn unevenness of the defect.

The present invention is configured as the second means. By comparing the individual spectrogram with the reference spectrogram formed as described above, the height of the individual spectrogram exceeding the reference spectrogram is determined. The point is to display the excess portion of the channel, and thereby detect thread unevenness as a defect.

Further, the present invention is configured as a third means, that is, on a display or a print paper,
The point is that the excess portion of the excess channel is displayed in different colors according to the excess amount.

[0015]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.

First, a plurality of samples sampled from the yarn to be measured are made to travel between the electrodes, and a signal proportional to the weight fluctuation of each sample is taken out there. The unevenness component of the wavelength is classified for each channel, and the individual spectrogram in which the fluctuation rate of each unevenness component is displayed by the height of the channel is formed. This point is similar to the conventional method, and FIG. 2 (A) is an analog waveform of a signal recorded by the above-mentioned applicant's trade name Evenness Tester, FIG. 2 (B).
Shows an individual spectrogram recorded by the same evenness tester.

Therefore, k individual spectrograms CH _n can be obtained from a plurality of samples, for example, at least 5, usually 10, and in many cases 20 or more samples.

Therefore, in order to see a problem common to all samples, k individual spectrograms CH _n (n = 1,
2, ... K) are averaged to form an overall spectrogram CH _OA as shown in FIG. That is, the k individual spectrograms CH _n are calculated for each channel L based on the following mathematical formula [Equation 1], and this result is collectively expressed to obtain the overall spectrogram C.
_HOA .

[0019]

[Equation 1]

Subsequently, a reference spectrogram CH _RF as shown in FIG. 3B is created from the overall spectrogram CH _OA as follows.

For each channel CH _{OA (L)} of the overall spectrogram, when it is higher than any of the adjacent channels on the left and right, it is corrected to the average value of the left and right channels. That is, the height of the channel to be corrected is the mean value of the left and right channels. However, the leftmost channel is compared only with the right channel, and the rightmost channel is compared only with the left channel and corrected to the average value. When the correction to this average value is performed from the leftmost channel (channel showing the shortest wavelength) to the predetermined channel on the right side of the overall spectrogram, there is a peak height channel (in the case of the example, wavelength 5 cm Channel) is formed, and the channel on the right side of the peak gradually decreases in height. Theoretically, assuming that the average fiber length contained in the yarn sample is LF, the peak wavelength is 2.5 to 3 times the LF, and therefore 4
The average value may be corrected up to the right channel showing a long wavelength of LF or more.

Therefore, the common ratio is calculated by considering the monotonous decrease in the height from the peak channel to the channel on the right side described above as a geometric reduction.

Then, the channel showing a long wavelength of 4 × LF or more is corrected to a value calculated by the common ratio obtained above.

After that, the reference spectrogram CH _RF is formed by combining the above-mentioned corrections for each channel CH _{OA (L)} of the overall spectrogram.

Therefore, the reference spectrogram CH _RF approaches the ideal spectrum as shown in FIG. 1C from the overall spectrogram CH _OA , but the ideal spectrum is theoretically a perfect ideal. On the other hand, the reference spectrogram CH _RF is a spectrogram as a realistic target obtained by removing only the abnormal periodic defects from the overall spectrogram CH _OA obtained from the plurality of measured samples.

Therefore, whether or not the measured sample has a realistic periodic defect that should be considered a problem is determined by comparing the overall spectrogram CH _OA with the reference spectrogram CH _RF or by the individual spectrogram CH _n. If there is a height portion exceeding the reference spectrogram CH _{RF (L)} for each channel, it can be judged that it is an abnormal periodic defect (uneven defect addition).

(First Embodiment) In the first embodiment of the present invention, the defective speckle unevenness is compared between the reference spectrogram CH _RF and the overall spectrogram CH _OA , and the following formula [Equation 2] is applied to each channel L.
4A, the sum of the results is shown as the average addition unevenness spectrogram CH _{AD (L)} in the reference spectrogram CH, as shown in FIG. 4 (A).
It is done by superimposing it on _RF .

[0028]

[Equation 2]

In FIG. 4 (A), the average addition unevenness spectrogram CH _{AD (L)} is shown superimposed on the reference spectrogram CH _RF , but the average addition unevenness spectrogram CH _AD exceeds the reference spectrogram CH _{RF .} Only _(L) may be extracted and displayed.

(Second Embodiment) In the second embodiment of the present invention, the defective speckled unevenness is referred to as a reference spectrogram CH _RF and an individual spectrogram CH _n (n = 1,
2, ..., K), and for each channel L, the calculation is performed based on the following formula [Equation 3], and the total of these results is k as shown in FIG. 4 (B). of particular additional unevenness spectrogram _{CH 1AD, CH 2AD, ···,} CH
_It is done by expressing it as _kAD .

[0031]

[Equation 3]

[0032] In FIG. 4 (B), k pieces of particular additional unevenness spectrogram CH _1AD, CH _2AD, · · ·,
Although CH _kAD is represented by itself, these k individual additional unevenness spectrograms may be represented by being superposed on the reference spectrogram CH _RF .

The above-mentioned average addition unevenness spectrogram CH
_{AD (L)} and / or individual addition uneven spectrogram CH
_{The kAD} is displayed on a display of a computer or printed on a printing paper by a printer, but it is preferable that the excess portion of the excess channel is color-coded according to the excess amount. For example, CH _{AD (L)} obtained by the formula [Formula 2] or C obtained by the formula [Formula 3]
Depending on the value of H _{n (L)} , when the value is 5 or more, it is gray, when it is 4 or more and less than 5, it is red, and when it is 3 or more and less than 4, it is yellow, 2
When the number is 3 or less, the color can be classified into blue, and when the number is 1 or more and less than 2, the color can be classified into white, whereby the display of the defective addition unevenness can be revealed.

It should be noted, and the case of displaying by superimposing the average added unevenness spectrogram CH _{AD (L)} as shown in FIG. 4 (A) on the reference spectrogram, k pieces of particular additional unevenness spectrogram CH _1AD, CH _2AD, ..., C
When H _kAD is superimposed and displayed on the reference spectrogram, it may be displayed with the ideal spectrum added at the same time.

[0035]

According to the present invention, an overall spectrogram CH _OA is formed by averaging the k individual spectrograms CH _n obtained corresponding to the number of samples, and the overall spectrogram C
A reference spectrogram CH _RF is formed by averaging the heights of adjacent channels of H _OA ,
For this reference spectrum Lot grams CH _RF, by comparing the overall spectrogram CH _OA or individual spectrograms CH _n, average added unevenness spectrogram CH _{AD (L)} or particular additional unevenness spectrogram that exceed the reference spectrum Lot grams CH _RF for each channel Since CH _kAD is displayed, it is possible to immediately and immediately judge abnormal defective addition irregularities.

With respect to this point, as described above, for example, when the additional unevenness is detected with reference to the theoretically obtained ideal spectrum, all the additional unevenness within the allowable I index range are displayed, and the yarn concerned is displayed. While it is difficult to detect the target realistic defect part for C, according to the present invention, k individual spectrograms C
Overall spectrogram CH _OA obtained by averaging H _n to reference spectrogram C
H _RF and forms the reference spectrogram CH _RF
Since the addition unevenness is detected by using as a standard, it is extremely advantageous in that only the abnormal abnormal addition unevenness that is a realistic target can be easily detected while discarding the allowable addition unevenness.

Then, the average addition unevenness spectrogram CH _{AD (L)} or the individual addition unevenness spectrogram CH
_If the excess portion of the excess channel of _kAD is color-coded and displayed on the display or the print paper according to the excess amount, it is possible to more easily visually recognize the target defective addition irregularity.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of a yarn unevenness component, (A) is an example showing that the yarn unevenness (solid line) includes a plurality of wavelength components (A, B, C), and (B) is formed by frequency analysis. FIG. 3C is an example of the individual spectrogram in which the ideal spectrum is added.

FIG. 2 illustrates yarn unevenness measured from one sample according to the present invention. (A) is an example of an analog waveform of a signal proportional to a weight change of a running sample, and (B) is the signal. It is an example of the individual spectrogram formed by performing the frequency analysis of.

FIG. 3 shows an overall spectrogram and a reference spectrogram according to the present invention,
(A) is an example of an overall spectrogram formed from k individual spectrograms, and (B) is an example of a reference spectrogram formed from the overall spectrograms.

FIG. 4 is a spectrogram of defect addition unevenness displayed by the present invention, (A) is an example of a display method in which an average addition unevenness spectrogram is superimposed on a reference spectrogram, and (B) is an individual addition unevenness. It is an example of a display method in which a spectrogram is represented alone.

─────────────────────────────────────────────────── ─── Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) D01H 13/26

Claims (3)

(57) [Claims] 1. A yarn unevenness component having different long and short wavelengths is obtained by running a plurality of samples sampled from a yarn to be measured, extracting a signal proportional to the weight fluctuation of each sample, and then performing frequency analysis or wavelength analysis from the signal. In the yarn unevenness detection method in which the individual spectrograms are classified according to the channels and the fluctuation rate of each yarn unevenness component is displayed by the height of the channel, k individual spectrograms obtained corresponding to the number of the samples are obtained. An overall spectrogram is formed by averaging, a reference spectrogram is formed by averaging the heights of adjacent channels of the overall spectrogram, and the overall spectrogram is compared by comparing the reference spectrogram with the overall spectrogram. Yarn unevenness detecting method characterized by displaying the height ultra excessive portion of the channel of which exceed the reference spectrogram of programs as average added unevenness spectrogram. 2. A yarn unevenness component having different wavelengths in terms of short and long wavelengths is obtained by running a plurality of samples sampled from a yarn to be measured, extracting a signal proportional to the weight fluctuation of each sample, and then performing frequency analysis or wavelength analysis from the signal. In the yarn unevenness detection method in which the individual spectrograms are classified according to the channels and the fluctuation rate of each yarn unevenness component is displayed by the height of the channel, k individual spectrograms obtained corresponding to the number of the samples are obtained. An overall spectrogram is formed by averaging, and a reference spectrogram is formed by averaging the heights of adjacent channels of the overall spectrogram, and the individual spectrogram is compared by comparing the reference spectrogram and the individual spectrogram. Out of riff Yarn unevenness detecting method characterized by displaying the ultra excessive portion of the channel height that exceeds the Reference spectrogram as a separate additional unevenness spectrogram. 3. The yarn unevenness detecting method according to claim 1, wherein the excess portion of the excess channel is color-coded and displayed according to the excess amount on a display or a print paper.